H. Houe, J. C. Baker, R. K. Maes, H. Wuryastuti, R. Wasito, P. L. Ruegg, J. W. Lloyd

J Vet Diagn Invest 7:321-326 (1995) Prevalence of cattle persistently infected with bovine viral diarrhea virus in 20 dairy herds in two counties in central Michigan and comparison of prevalence of antibody-positive cattle among herds with different infection and vaccination status H. Houe, J. C. Baker, R. K. Maes, H. Wuryastuti, R. Wasito, P. L. Ruegg, J. W. Lloyd Abstract. All cattle in 20 dairy herds randomly selected from herds participating in the Dairy Herd Improvement Association program in 2 counties in central Michigan were tested for the presence of bovine viral diarrhea virus (BVDV). Virus-positive animals were retested to ascertain persistent infection with the virus. A total of 5,481 animals were tested for presence of BVDV. In 9 of the herds, all animals were also tested for virus neutralizing antibody titer. Based on infection and vaccination status, these 9 herds were divided into 3 different herd categories: A, 5 herds with currently no cattle persistently infected (PI) with BVDV and without any vaccination program against BVDV in recent years; B, 2 herds with no current PI cattle but using killed BVDV vaccines; and C, 2 herds with PI cattle. PI cattle were detected in 3 out of 20 herds (15%). A total of 7 of 5,481 animals (0.13%) were PI. The mean prevalences of antibody carriers in herd categories A, B, and C were 28.8%, 76.4% and 90.6%, respectively. For one herd in category A, antibody analyses indicated that mostly young stock was seropositive, suggested recent BVDV infection in a previously closed and naive herd. Cattle in category B herds were vaccinated with killed vaccine from the age of 15 months. These herds had several antibody negative animals among the younger cows, suggesting incomplete protection against BVDV infection. In the 3 herds in which PI animals were detected, all cattle had been vaccinated with killed vaccine. The antibody-positive animals had antibody titers that were significantly different both among herds and among herd categories. The antibody titers of animals exposed to PI animals were significantly higher than those of animals vaccinated with killed vaccine. munocompetence) induces specific immunotolerance to BVDV. 13 Such calves are born persistently infected (PI) and shed large amounts of BVDV for life. PI cattle play a key role in the epidemiology of this disease complex because they are important excretors of virus. Furthermore, they can develop mucosal disease, either by mutation of the existing persistent ncp biotype of the virus to the cp biotype or by super-infection from an outside source of the cp biotype. 2,3,11 Determination of the prevalence of PI cattle is therefore an important tool in the epidemiological study of BVDV because it Infection with bovine viral diarrhea virus (BVDV) was described for the first time in 1946 in the United States. 16 BVDV is member of the genus Pestivirus in the family Flaviviridae. 7 Based on its growth in cell culture, BVDV biotypes can be designated as cytopathogenic (cp) or noncytopathogenic (ncp). 8,12 Postnatal infection with ncp BVDV (acute infection) is most often subclinical, but temporary symptoms such as fever, inappetence, depression, and reduced milk pro- duction may be seen. In pregnant animals, however, fetal infection with BVDV may result in repeat breeding, abortion, congenital defects, and growth retardation. 5,17,19 Fetal infection with ncp BVDV prior to day 90-120 of gestation (i.e., before development of im- is an indirect measure of the number of fetal infections occurring in early pregnancy. In areas without BVDV vaccination, the prevalence of antibody carriers has been used to estimate the incidence of infection 10 and to predict presence or absence of PI cattle. 9 However, From the Department of Clinical Studies, The Royal Veterinary the distribution of antibody carriers in areas where and Agricultural University, Bülowsvej 13, DK-1870 Frederiksberg vaccination is available needs further elucidation. C, Denmark (Houe), the Department of Microbiology, Animal Health The objectives of this study were to 1) estimate the Diagnostic Laboratory (Maes), the Department of Large Animal prevalence of PI animals among dairy cattle in herds Clinical Sciences (Baker, Ruegg, Lloyd), College of Veterinary Med- enrolled in the Dairy Herd Improvement Association icine, Michigan State University, East Lansing, MI 48824-1314, and the School of Veterinary Medicine, Gadjah Mada University, Yo- (DHIA) in 2 counties in central Michigan and 2) ingyakarta, Indonesia (Wuryastuti, Wasito). vestigate the prevalence of antibody-positive animals Received for publication November 4, 1994. in herds with different infection and vaccination status. 321

322 Houe et al. Materials and methods Selection of dairy herds. The sampling frame for this study consisted of all herds (n = 118) enrolled in DHIA in Ionia and Clinton counties, which are ranked number 3 and number 6, respectively, for the number of dairy cows per county in Michigan. A table of random numbers was used to select herds that were solicited to participate in the study. A letter describing the study was sent to owners of selected herds. Herds whose owners declined to participate in the study were replaced by additional randomly selected herds until owners of a total of 20 herds agreed to participate. Blood sampling. Blood samples from all cattle were collected in Vacutainer tubes without anticoagulant. In 2 herds, samples were not collected from some of the young stock because of lack of available restraints, i.e., 66 animals in herd 8 were on pasture and 50 animals in herd 9 were steers. Animals that were virus positive were retested at a minimum of 3 wk after the first sample. Animals were rated as PI if they tested virus positive on both occasions or if they tested positive on the first occasion and were subsequently removed from the herd because of a disease course concordant with persistent infection before the second sample could be obtained. Virological examination. All sera were examined for BVDV by an immunoperoxidase monolayer assay (IPMA) similar to the procedure described previously. 14 Microtiter plates were seeded with bovine turbinate cells and incubated for 24 hr at 37 C in a humidified 5% CO 2 atmosphere. The wells were then inoculated with 15 µl of undiluted serum from each animal. The inoculated plates were reincubated for 72-96 hr. The plates were then drained and rinsed twice with 0.01 M phosphate-buffered saline (PBS). Cells were fixed with 35% acetone in PBS containing 0.02% bovine serum albumin and incubated for 10 min at room temperature. The cells were then dried at 37 C for 90 min or until completely dry. One hundred microliters of polyclonal BVDV antiserum was added to each well and incubated for 30 min at room temperature. Plates were then drained and rinsed 3 times with wash buffer (PBS containing 0.05% Tween 20), and 50 µl of protein G-horseradish peroxidase diluted 1:1,000 with binding buffer (PBS containing 0.05% Tween 20 and 2.95% NaCl) were added to each well and incubated for 15 min at room temperature. Plates were then drained and washed twice with wash buffer, and 100 µl of freshly prepared substratechromogen solution (3-amino-9-ethylcarbazole dissolved in N,N-dimethyl-formamide in sodium acetate with hydrogen peroxide) was added to each well and incubated in a dark place for 1 hr at room temperature. Finally, the plates were drained and rinsed twice in tap water. Virus-positive cells have brownish-staining cytoplasm. Sera that were virus positive by IPMA and negative control sera from the same herds were also tested by the fluorescent antibody technique following standard procedures. Virus neutralization (VN) test. Sera from all animals in 9 herds were examined for VN antibody titer against BVDV. Based on the presence of PI animals and the use of vaccination, the herds were selected to represent the following three categories: A, 5 herds with no PI cattle and that had not been vaccinated with BVDV vaccines (in 3 of these herds, no BVDV vaccine had been used for more than 5 years, and in 2 herds, nos. 5 and 12, BVDV vaccine had been used until about 4 years earlier); B, 2 herds with no PI cattle but vaccinated with killed BVDV vaccines; C, 2 herds with PI cattle. Herds in category C also had been vaccinated with killed BVDV vaccines. Further, the herds were selected among those herds with good recording of birth dates of the animals. VN antibody titers were determined according to previously described methods. 4 The sera were inactivated at 56 C for 30 min. Two-fold dilutions of each serum, ranging from 1:4 to 1:4,096, were tested. The amount of virus used per test was 500 TCID 50 of the cytopathic Singer strain. The microtiter system consisted of 15,000 bovine turbinate cells/ well. Each test included a back titration of the virus and a positive and negative serum control. The antibody titer was read as the highest dilution with complete inhibition of cytopathic effect. Collection of herd data. Every herd owner was asked to fill out a questionnaire reporting information on pasturing, purchase of cattle within the last 3 years, possible contact with animals from other herds, and use of BVDV vaccines. Statistical analysis. Differences in antibody prevalences were tested by the chi-square test, and differences in antibody titers among herds were tested by the Kruskal-Wallis test (chi-square approximation), using PROC NPAR1WAY in SAS. 18 Results Virological examination. Seven animals were virus positive with the first blood sample. Only 6 animals were available for retesting, and they remained virus positive. One viremic animal in herd 6 could not be retested because it died due to respiratory disease before a second serum sample could be obtained. This animal was also classified as PI. Thus, the total number of PI animals was 7, corresponding to an overall prevalence of PI animals of 0.13%. PI animals were found in 3 of 20 herds examined, thus giving a herd prevalence of 15% (Table 1). All virus-positive sera also tested virus positive with the fluorescent antibody technique. In addition, 1 serum sample stained weakly in the IPMA, indicating a possibly virus-positive animal. Unfortunately, this animal could not be retested because it only had a metal ear tag and the farmer objected to catching and restraining all animals again to look for metal tag numbers. However, when testing all animals (n = 21) in the same group the possibly virus-positive animal was part of, the mean and medium antibody titers were 285 and 128, respectively. Because these titers were comparable to titers in herds using killed vaccine, the suspected virus-positive animal was considered not PI. Prevalence of seropositive animals. The prevalences of seropositive animals in categories A-C were 28.8%, 76.4%, and 90.6%, respectively (Table 2). Antibody prevalences varied significantly among the 9 herds (P

Cattle persistently infected with BVDV 323 Table 1. Prevalence of animals persistently infected with bovine viral diarrhea virus in 20 randomly selected dairy herds in 2 counties in Michigan. Table 2. Antibody prevalence in 9 herds in which all animals had been tested for bovine viral diarrhea virus and antibodies. < 0.001, chi-square) and among the 3 herd categories (P < 0.001, chi-square). In herd category A, the prevalence of seropositive animals varied among herds (0-82%, Table 2). Herd 13 had no seropositive animals. Only 2 bulls had been added to this herd in the last 3 years. Animals were not pastured in this herd. Herd 3 had 6 antibodypositive animals. Two of the antibody-positive animals were between 9 and 18 months old, which indicates recent infection of the herd. Only 1 animal had been purchased during the last 3 years, and this animal was antibody negative. The animals were pastured but were separated from other pastured animals originating from other herds by 0.8 km. The owner reported that while on pasture there was potential for contact with deer and that 2 years prior a cattle trailer had been used to transport other animals. Herd 11 had only 1 seropositive animal among cows (8 years old), but several of the young stock were seropositive (Fig. 1). Most animals had low titers (median titer = 4). The cattle were not pastured, and no animals had been purchased within the last 3 years. The owner could not remember his cattle having any contact with other cattle, except for 3 or 4 cows that had gone to a veterinary clinic for surgery. In herd 12 all animals more than 3 years of age were antibody positive, and most animals had high antibody titers (median titer = 1,024). All animals between 9 and 18 months of age were antibody negative (Fig. 2), and several animals had been added to the herd in recent years. Herd 5 had the highest prevalence of antibody carriers (82%) in category A. Many animals had been added to this herd in recent years, but the vaccination status of added animals was not known. All animals in this herd more than 4 years of age were antibody positive (Fig. 3). In herd category B, the prevalence of antibody carriers was 75-77% in the 2 herds. The antibody titers were medium high (median = 16 and 32, respectively). Most seronegative animals were found among young stock, but several younger cows were also seronegative (Figs. 4, 5). In herd category C, the antibody prevalences were 96% and 76%, respectively. The seropositive animals had high titers (median = 1,024 and 512). Among the 6 PI animals in these 2 herds, 3 PI animals were seronegative, 2 PI animals had antibody titers of 4, and 1 PI animal (3 months old) had an antibody titer of 8. In herd 15, all but 2 young animals were seropositive, whereas 22 cows were antibody negative. The PI calf had been removed from the cow barn when it was 1 week of age and had been kept in a building 50 yards from the cows. At the time this herd was tested, the PI calf was 6 months old. Antibody titers among herds. Antibody titers showed a high degree of variation among herds (Table 2). For only seropositive animals, the antibody titers varied

324 Houe et al. Figure 1. Herd 11. Age distribution of carriers of antibodies against bovine viral diarrhea virus. significantly among herds all 9 herds (P = 0.0001, Kruskal-Wallis). Also, antibody titers among herd categories varied significantly (P = 0.0001). When comparing antibody titers among herds within each herd category, only herds in categories A and B varied significantly: A, P = 0.0001; B, P = 0.0005; C, P = 0.71. The antibody-positive animals in herd 12 showed antibody titers similar to those of the 2 herds in category C, i.e., differences in antibody titers among herds 12, 6, and 15 were not significant (P = 0.87). Discussion The prevalence of PI animals (0.13%) and PI herds (15%) in this study is low compared to that reported Figure 2. Herd 12. Age distribution of carriers of antibodies against bovine viral diarrhea virus. Figure 3. Herd 5. Age distribution of carriers of antibodies against bovine viral diarrhea virus. in other studies. 15 However, many previous studies are more or less biased towards clinical outbreaks or examination of certain age groups. The herds included in the present study were all enrolled in the DHIA program, which only includes approximately 50% of the herds in 2 counties of Michigan. The herds were therefore selected from among those with better management, which could mean better control of the infection. During a study in Denmark, 19 herds representative of the cattle population were investigated; 1.4% of the animals tested were PI, and PI animals were found in 53% of the herds tested. In herds without PI animals, 43% (14-67%) of the animals were antibody positive. 10 In the present study, antibody analyses of category A herds revealed an antibody prevalence of 28%. However, excluding herd 5, which may have had many seropositive animals due to addition of cattle from vaccinated herds, the prevalence of antibody carriers was 23% (0-53%). The antibody analyses thus confirms a lower infection rate in the herds in Michigan. The antibody analyses also shows that it is possible to obtain a totally naive herd in an infected area (herd 13). However, this is often a dangerous situation, because herds with similar precautions against reintroduction of infection revealed recent infection (herds 3, 11). The cause of past infection in these 2 herds (deer, cattle trailer, animals sent to a veterinary clinic) is not certain. In herd 11, many of the young stock had been infected. If 1 of these had been a heifer in early pregnancy, the result probably would have been a PI calf. Because all cows except 1 were still antibody negative in this herd, the presence of a PI calf would over time cause many infections and seroconversions among the

Cattle persistently infected with BVDV 325 Figure 4. Herd 1. Age distribution of carriers of antibodies against Figure 5. Herd 14. Age distribution of carriers of antibodies bovine viral diarrhea virus. against bovine viral diarrhea virus. pregnant cows and hence the possibility of numerous PI calves 6-10 months later. Herds with numerous PI animals have been discovered in the United States. In a survey of 3,157 animals among 66 nonrandomly selected herds in the USA, 54 cattle (1.7%) were PI. These PI animals came from 2 different herds, 1 of which had a history of BVDV infection. Another 4 herds contained a total of 6 viremic animals, but retesting to confirm PI was not possib1e. 1 Therefore there seems to be a low herd prevalence, but occasionally an infected herd may have numerous PI animals. Selection of such a herd in our Michigan study would significantly increase the overall prevalence of PI animals. Because such a herd was not randomly selected, the overall prevalence of PI animals found in this study might be an underestimation. Presence of only seropositive animals among animals above a certain age, as in herd 12, may be indicative of a previous exposure to a PI animal that had later been culled from the herd. This is further supported by the high antibody titers, which were similar to titers of animals in herd category C. Antibody analyses of herds in category B suggest that the protection, as assessed by presence of antibody titers, afforded by killed vaccine in these 2 herds may have been incomplete. However, because the number of PI animals in infected herds was low, the vaccine probably offered protection against fetal infection in most cases. In herd category C, there was a high prevalence of antibody carriers. Furthermore, the antibody titers were high. Thus the presence of both high prevalence of seropositive animals and high antibody titers might be a useful indication of presence of PI cattle in the group. Ninety-six percent of the animals in herd 6 were antibody carriers, even though the animals were housed on 3 different farms. But this herd had several PI animals and there was at least 1 PI animal on each farm. Presence of several seronegative cows in herd 15 with only 1 PI animal indicates that housing of animals in separate groups may delay the spread of infection from the PI animal. Despite the possibility for an underestimation because only herds enrolled in DHIA were selected for the study, the prevalence is still low compared with the prevalence study in Denmark. Some PI animals might have remained undetected because colostral antibodies among young calves could have interfered with the IPMA. However, the PI animal in herd 9 was only 1 week old when first detected. The IPMA used in the Danish study was also performed on serum for all animals, and the testing protocols are therefore comparable in the 2 studies. There are in general a number of important epidemiological differences between the 2 areas, most of which are in favor of a lower prevalence in Michigan. Lower overall concentration of animals, less pasturing of cattle, and use of vaccine may be factors reducing the risk of transmission of infection among herds in Michigan. Although not addressed in this study, the use of live vaccines containing cp BVDV on some occasions within the general population may induce mucosal disease and thereby reduce the number of PI animals. A cytopathic virus isolated from a mucosal disease case was closely related to a vaccine strain. 6 It was not possible from this study to evaluate the overall beneficial effect of vaccines. Further, it is not known in detail whether vaccination was performed according to label instructions on all occasions. Pres-

326 Houe et al. ence of PI animals in 3 herds and evidence of infection within recent years in another 4 herds (herds 3, 5, 11, 12) shows that use of vaccines has not eliminated circulation of BVDV in the area. However, because PI animals only occurred in low numbers in the infected herds, disease problems due to BVDV infections probably would increase significantly if vaccinations were withdrawn. Acknowledgements We thank Alice Murphy and Leslie Tengelsen for excellent technical assistance. The Michigan Agricultural Experiment Station supported this study. Financial support came from 1433 Animal Health Funds and the Danish Agricultural and Veterinary Research Council. Dr. S. Bolin, National Animal Disease Center, Ames, IA, provided the polyclonal BVDV antiserum. References 1. Bolin SR, McClurkin AW, Coria MF: 1985, Frequency of persistent bovine viral diarrhea virus infection in selected cattle herds. Am J Vet Res 46:2385-2387. 2. Bolin SR, McClurkin AW, Cutlip RC, Coria MF: 1985, Response of cattle persistently infected with noncytopathic bovine viral diarrhea virus to vaccination for bovine viral diarrhea and to subsequent challenge exposure with cytopathic bovine viral diarrhea virus. Am J Vet Res 46:2467-2470. 3. Brownlie J, Clarke MC, Howard CJ: 1984, Experimental production of fatal mucosal disease in cattle. Vet Rec 114:535-536. 4. Carbrey EA, Brown LN, Chow TL, et al.: 1971, Recommended standard laboratory techniques for diagnosing infectious bovine rhinotracheitis, bovine virus diarrhea and shipping fever (parainfluenza-3). Proc US Anim Health Assoc 75:629-648. 5. Done JT, Terlecki S, Richardson C, et al.: 1980, Bovine virus diarrhoea-mucosal disease virus: pathogenicity for the fetal calf following maternal infection. Vet Rec 106:473479. 6. Donis RO, Krejci A: 1993, Fatal iatrogenic mucosal disease caused by modified live bovine viral diarrhea virus? Proc Symp Pestiviruses 2: 149-155. Francki RIB, Fauquet CM, Knudson DL, Brown F: 1991, Classification and nomenclature of viruses. Fifth report of the International Committee on the Taxonomy of Viruses. Arch Virol Suppl 2:228-229. Gillespie JH, Baker JA, McEntee K: 1960, A cytopathogenic strain of virus diarrhea virus. Cornell Vet 50:73-79. Houe H: 1992, Serological analysis of a small herd sample to predict presence or absence of animals persistently infected with bovine virus diarrhoea virus (BVDV) in dairy herds. Res Vet Sci 53:320-323. Houe H, Meyling A: 1991, Prevalence of bovine virus diarrhoea (BVD) in 19 Danish dairy herds and estimation of incidence of infection in early pregnancy. Prev Vet Med 11:9-16. Howard CJ, Brownlie J, Clarke MC: 1987, Comparison by the neutralisation assay of pairs of non-cytopathogenic and cytopathogenic strains of bovine virus diarrhoea virus isolated from cases of mucosal disease. Vet Microbiol 13:361-369. Lee KM, Gillespie JH: 1957, Propagation of virus diarrhea virus of cattle in tissue culture. Am J Vet Res 18:952-953. McClurkin AW, Littledike ET, Cutlip RC, et al.: 1984, Production of cattle immunotolerant to bovine viral diarrhea virus. Can J Comp Med 48:156-161. Meyling A: 1984, Detection of BVD virus in viremic cattle by an indirect immunoperoxidase technique. In: Recent advances in virus diagnosis, ed. McNulty MS, MacFerran JB, pp. 37-46. Martinus Nijhoff, Boston, MA. Meyling A, Houe H, Jensen AM: 1990, Epidemiology of bovine virus diarrhoea virus. Rec Sci Tech Off Int Epizoot 9:75-93. Olafson P, MacCallum AD, Fox FH: 1946, An apparently new transmissible disease of cattle. Cornell Vet 36:205-213. Roeder PL, Jeffrey M, Cranwell MP: 1986, Pestivirus fetopathogenicity in cattle: changing sequelae with fetal maturation. Vet Rec 118:44-48. SAS: 0000, SAS user s guide. SAS Institute, Cary, NC. Virakul P, Fahning ML, Joo HS, Zemjanis R: 1988, Fertility of cows challenged with a cytopathic strain of bovine viral diarrhea virus during an outbreak of spontaneous infection with a noncytopathic strain. Theriogenology 29:441-449.